NVIDIA GeForce 6800GT Roundup
Test configuration
Since PCI-Express x16 video cards take part in the tests, we activated our both test benches - both AGP and PCI-E.
| Test configuration |
PCI-E |
AGP |
| CPU |
P4 3.6Mhz 800FSB LGA |
P4 3.2Mhz 800FSB (Northwood D1) |
| Mb |
Intel D915GUX (i915G) |
Epox 4PDA2+ (i865PE) |
| Memory |
PC2-4300 (533MHz DDR2) 2x256Mb in the dual-channel-mode
Memory latency timings - 4:4:4 |
PC2700 2x256Mb 400Mhz in the dual-channel mode
Latency timings - 2.5-8-4-4 |
| HDD |
Maxtor MaxlineIII SATA, 2x250Gb Matrix RAID |
IBM, 120Gb |
| CD-ROM |
ASUS CR/R-RW 32 |
MSI XA52P COMBO Writer SATA |
| OS |
WinXP + SP2 + DirectX 9.0c |
WinXP + SP2 + DirectX 9.0c |
| Drivers |
ForceWare 65.76
Catalyst 4.9 WHQL |
ForceWare 65.76
Catalyst 4.9 WHQL |
Test software:
Synthetic benchmarks:
- 3DMark2005 Pro Patch 110;
- 3DMark2003 Pro Patch 340;
- 3DMark2001SE Patch 330;
- Codecreatures v1.0.0 (a DirectX 8.1 application, shaders on, Hardware T&L);
Gaming benchmarks:
- Unreal Tournament 2003 (Direct3D, Hardware T&L, vertex shaders, Dot3, cube texturing.);
- Unreal Tournament 2004 (Direct3D, Hardware T&L, vertex shaders, Dot3, cube texturing.);
- AquaMark 3 (DirectX 9.0, Vertex Shaders 1.1/1.4/2.0, Pixel Shaders 1.1/1.4/2.0, Hardware T&L, AquaMark3 Triscore mode);
- HALO: Combat Evolved 1.2 (DirectX 9.0, Vertex Shaders 1.1/1.4/2.0, Pixel Shaders 1.1/1.4/2.0, Hardware T&L, quality set to the maximum possible);
- FarCry ver1.1 (DirectX 9.0, Pixel Shaders 2.0, our own demo "3Dnews001" was used).
- DOOM 3(timedemo demo1.demo).
As usual, we overclocked the cards with RivaTuner.
Tests of video cards on the base of the GeForce 6800GT chip
For completeness of the picture, we'll present results for both AGP and PCI-E cards albeit measured on different benches. In these difficult conditions of transition period (which has been lasting for 5 years by now :) we can deduce an approximate picture of results produced for all the resolutions:
That is, it's absolutely impossible to make head or tails in here but for the general tendency of the very old UT2003 engine towards the processor dependence. In our tests, this is the only "old game - we'll have to give them up, like 3DMark2001SE.
With new games like Far Cry or DOOM3 there is no such picture, and the results produced at different AGP and PCI-E benches can be regarded more or less relevant, but over there the graphs merge into a unified pool in which it is impossible to figure anything out. So we decided to show the alignment of forces for one resolutions.
A graph for 3DMark 2005Pro taken only on the AGP bench with today's cards participating. The interest to this new test is so high that we decided to take results for all the cards supporting PS2.0 and PS3.0. The benches were maximally approached to one another at performance, with the latest ATI's and NVIDIA's recommended drivers used. In the end, we produced the following picture:
Interestingly, X800XT Platinum Edition on the AGP bench consistently showed lower result than the regular X800XT on PCI-E. We racked our brains for a long time to finally arrive at the conclusion that it was just the case where the PCI-E bus already starts demonstrating its advantages at the bandwidth. Anyway, the test is positioned as "a benchmark of tomorrow".
 |
Top Stories: |
 |
 |
|
MoBo:
| |
|
|
VGA Card:
|
|
|
|
CPU & Memory:
|
|
